The past few decades have been rough on the Midwest. What was once
America's dominant manufacturing region, and one of its most
populous, has taken a back seat to more vital areas of the country.
While the situation might not be as dire as it's often portrayed,
there is some truth to that grim picture. As warmer parts of the
country have exploded in population, the Midwest has languished by
comparison, losing representation in Congress. Worse, Midwestern
economic growth has mirrored its population trend. Over the past 10
years, every state in the Midwest has had a lower rate of output
growth than the nation as a whole. Globalization has made
manufacturing cheaper in other parts of the world, and new
technologies have reduced the significance of natural advantages like
the Great Lakes.
There is, however, one resource that reigns supreme in the "knowledge
economy," and the Midwest is rich with it: educated people. That's
thanks to years of prosperity supporting an extensive network of
universities, particularly the large land-grant research institutions
that constitute the Big Ten.
But that invites the question: If educated workers are so abundant in
the Midwest, why hasn't the problem of slower economic growth fixed
itself? To understand this challenge, representatives from Midwestern
universities, state and local governments, and nongovernment
organizations held a conference June 26-27 at the Federal Reserve Bank
of Minneapolis about the future of the region's economy.
The conference was organized by the Center on Institutional
Cooperation, a collaborative of the Big Ten universities plus the
University of Chicago. The CIC is celebrating its 50th anniversary in
2008, making it the oldest and largest such institution in the
country. Representatives from the Chicago Fed, which hosted two
similarly themed conferences in 2005 and 2006, were also in
The conference focused on the role of human capital—the skills,
knowledge and social aptitudes of workers and entrepreneurs—in
The conference kicked off Thursday evening with a reception, dinner
and words from Minneapolis Fed President Gary Stern and University of
Minnesota President Robert Bruininks on the institutional cooperation
between the bank and the university.
Following their remarks was a keynote address from Robert Lucas,
influential economic theorist, Nobel laureate and University of
Chicago professor. Lucas declared up front that he would not be
talking about regional economic cooperation. Instead he discussed the
bigger-picture issues of growth and the sources of development.
The next day began with Lance Lochner, of the University of Western
Ontario, speaking on "The Private Returns to Human Capital."
Lochner's talk drew on research he has done with James Heckman,
University of Chicago econometrician and Nobel laureate, on schooling
The social returns to schooling—reduced crime, higher tax revenues,
increased innovation and so on—have received a lot of attention.
Lochner focused instead on the benefits and costs to individuals,
since individuals ultimately must make their own educational choices.
Lochner's analysis considered those costs and benefits of schooling
that are easily measurable (such as tuition, forgone earnings,
increased future earnings) while staying silent on those that are not
(such as distaste for school or pride in learning).
A key problem with measuring the returns to schooling is that the
costs of education are mostly incurred up front, but the benefits
accrue over time. To estimate the returns to education, one therefore
needs to choose a particular discount rate for future benefits, and
that choice makes a big difference in the estimate of lifetime
Lochner's trick was to use a measure called the "internal rate of
return." This is the discount rate required for a person to be indifferent
between two lifetime income streams. For example, given potential
future earnings and wages that could be earned instead of going to
school, a young white male deciding not to finish high school would
have to be very impatient, as much as someone requiring an interest
rate of 46 percent to save.
Lochner's analysis ended with a puzzle. College enrollments and
graduations soared between 1940 and 2000, even though the financial
return to college completion increased only modestly. By contrast,
high school graduation rates have actually declined slightly in
recent decades, despite skyrocketing rates of return to finishing
Having explored what drives people to acquire human capital, Lochner
was followed by John Kennan of the University of Wisconsin-Madison,
who talked about what draws human capital across state borders.
Keenan started by reviewing the facts from the U.S. Census on
migration of the college-educated.
Contrary to popular belief about so-called brain drain in the
Midwest, exodus of college-educated workers from the region does not
appear to be much more severe than out-migration from states in other
parts of the country. Economically distressed Michigan, for example,
has about the same rate of retention of college graduates as Virginia
or Massachusetts. And while younger college-educated workers from the
Midwest are more likely to be working outside their home states than
the national average, as they enter their late 30s and beyond,
they're less likely, indicating they tend to return home later in
Keenan then presented a model of human capital migration. In the
model, workers with a given level of education earn different wages
in different places, but the cost of moving presents a barrier to
following the money. Keenan then took the model a step further,
incorporating data from the National Longitudinal Study of Youth to
calibrate it. His results showed that there really isn't all that
much migration among states, so the cost of migrating must be high
relative to the expected gains.
Midwestern newspapers are rife with stories about young people
fleeing their homes after college, so why this surprising finding? In
part it's because the data are messy, Keenan explained. While people
generally move toward areas with higher incomes, many also move in
the "wrong" direction, toward lower-income states. Further, many who
migrate do so numerous times, so there is a disconnect between the
amount of migration and the number of migrants. Overall, sustained
differences in incomes do cause migration, but it takes a long time
to materialize. Given this, Keenan suggested that policy changes that
affect migration shouldn't be assessed for at least five years after
the policies are in place.
Closing out the morning was Paul Glewwe, presenting on a topic near
and dear to the hearts of many academic administrators in attendance:
the contribution of research universities to state economies. Glewwe,
a professor in the University of Minnesota's applied economics
department, specifically focused on the contribution of his
university, along with the Minnesota State Colleges and Universities
System. In contrast to Lochner, Glewwe focused on public higher
education's public benefits, rather than the private ones.
To calculate those benefits, Glewwe and his co-author performed a
simple thought experiment. What would happen, they asked, if
Minnesota's public colleges had their funding withdrawn? Assuming
that tuition would rise to private levels, they estimated that a
tuition increase of $1,500 (in 1988 dollars) would lead to a decrease
in enrollment of 2.5 percentage points.
Many economists believe the public benefits to higher education
include the positive external or "spillover" effects that higher
wages for the educated have on the wages of others, increased voter
participation and reduced incarceration costs due to lower crime
rates. (It's also believed that since educated people may be
healthier on average, health care expenditures may be reduced, but
this effect is harder to detect.) Glewwe estimated the reduction in
these benefits from higher tuition, and just for good measure he also
looked at the loss in private benefits.
After all this was done, the current costs of state funding could be
compared to the lost benefits. Even using conservative estimates of
the wage spillover effects and ignoring the benefits of research by
public universities, Glewwe calculated a substantial benefit from
public higher education, with a benefit-cost ratio of about 2-to-1.
Still, the conference was a long way from answering the main question
at hand: What can be done about the Midwest's lagging economy?
Fighting Balkanization the right way
Over lunch, conference participants heard from Minneapolis Fed Senior
Vice President and Director of Research Art Rolnick about the
benefits of early childhood education. Rolnick presented his case
that high-quality preschool for at-risk children is possibly one of
the best investments state and local governments can make in economic
development. (For more on Rolnick's research with Minneapolis Fed
Associate Economist Rob Grunewald, see The Region, December 2003 Supplement.)
After lunch was the first of two panels on Midwestern development.
Richard Longworth, senior fellow at the Chicago Council on Global
Affairs, started the panel discussion with "The Balkanized Midwest
and What We Can Do about It," which was based on his recent book,
Caught in the Middle.
Longworth's theme was that while the Midwest is geographically and
culturally one region, it is politically fragmented by its multistate
structure, and this fragmentation prevents real movement forward. For
instance, Longworth noted that each of the Big Ten universities
represented in attendance has its own business, law and medical
schools—some excellent, others just decent. However, if those
universities could move beyond parochialism and pool their resources,
the Midwest could have several truly world-class institutions in each
field. Longworth ended with a call for greater cooperation by
Midwestern universities, perhaps along with the creation of a think
tank or "Midwestern studies" department, to help spur greater
economic and political cooperation across Midwestern states.
After Longworth finished, Tom Holmes, professor at the University of
Minnesota and research consultant at the Minneapolis Fed, provided an
economist's perspective on Longworth's talk. He started by reviewing
the economics of cooperation. Market competition usually induces
efficient outcomes from decentralized decisions, but there are two
types of market failures that may require collective action to
overcome: externalities and increasing returns.
As an example of externality, Holmes drew on a resource that is vital
to the region: the Great Lakes. If one state decides to allow more
dumping in a lake or divert water outside the watershed, the water
available to others is affected. Classic examples of increasing
returns include rail networks and power grids; they exhibit economies
of scale—double their size, and profits will more than double.
Railroads and power grids are also often shared by states, which is
crucial. Holmes cautioned that regional cooperation only makes sense
to the extent that one can clearly identify the market failure and,
in its resolution, include only those parties affected. It wouldn't
make sense to include Arizona in a Great Lakes water conservation
compact, and Puerto Rico probably doesn't want to get in on planning
a Midwestern railroad. So Holmes argued that before any regional
cooperation commences, the states involved should clearly identify
the market failure and "draw a circle around it."
One implication of this for the proposals Longworth put forward is
that while there probably aren't too many research universities in
the Big 10, there may be too many redundant academic departments
among them. The "research corridors" that have become popular with
states seeking to develop their biotech and other research-intensive
industries have not shown much evidence of producing collaborations
that wouldn't have happened otherwise, and the case for interstate
research corridors may be even weaker, due to distance alone. Holmes
also cautioned against "smokestack chasing" (cities or states trying
to compete with each other for existing businesses). While such
chasing might be a boost to a local economy, it doesn't create any
new jobs or prosperity in the aggregate, and can be a costly use of
public funds if tax rebates or subsidies are used as lures. (For more
on this issue, see the Minneapolis Fed's 1994 Annual Report essay,
"Congress Should End the Economic War Among the States," by Melvin
Burstein and Art Rolnick.")
Overall, Holmes said, while talk and collaboration are great,
regional decision-making in higher education and elsewhere should be
decentralized, unless the market failure and the parties it affects
can be clearly identified. What, then, can university, state and
local officials do to encourage regional development? That was the
topic of the day's final two sessions.
Beyond the ivory tower
First up was a panel from Chicago to talk about that city's
exceptional success. Frank Beal, from the Chicago Metropolis 2020
Project, talked about how the city was able to fight the trend of
Midwestern stagnation. The 2020 project brought together an
impressive collection of retired business leaders to move Chicago
into the 21st century, and Beal shared the lessons of their
The running themes in all the project's work are about keeping the
region economically competitive, focusing on the metropolitan area
for action—rather than the city or state level—and promoting greater
opportunities for the disadvantaged. This has led to an emphasis on
four main areas: the criminal justice system, early childhood
education, transportation and housing. Beal stressed that the project
has succeeded by keeping a narrow, well-defined focus (avoiding
"agenda creep"), concentrating on policy, being ambitious and not
getting bogged down in discussions and collaborations with other
The lessons for the CIC, Beal said, were clear. Echoing Holmes, he
said the CIC should make sure its focus is well-defined, but he also
said it should move its focus to policy and not simply institutional
cooperation. Following Beal were Connie Shoemake and Mark Cleverley
from IBM to discuss the role of technology in achieving those goals.
The final discussion was led by Michigan State University President
Lou Anna K. Simon. She kept her remarks brief, wanting to open the
floor to discussion. Simon stressed that in order to use
institutional cooperation as a tool for regional development, the
university officials present would have to maintain an ambitious
attitude toward affecting policy and better coordinate the use of
their existing assets. In the discussion that followed, conference
participants brainstormed ideas for future collaboration.
The conference offered a few important lessons for those future projects.
First, possibly the most effective way to ensure long-term economic
prosperity is through developing workforce human capital. Because
they come from universities, many conference participants are already
in this game, but it might help to start the push earlier, by
supporting early childhood education for at-risk kids. Given what
Lochner showed about the returns to high school graduation, this is
A second lesson: Don't worry too much about "brain drain." As Keenan
demonstrated, most educated workers stay in their home states for
much of their careers, and many who do leave eventually return to
start families. As important, regional development isn't the primary
goal of human capital policy; the goal is to provide workers with
greater opportunities, and that will lead to overall development. Any
regional development that stems from that, Lucas said, is incidental.
Finally, universities should appreciate the limits to what strategic
cooperation can accomplish. Not all collaborations make sense, and
Holmes' caution about bringing in only the affected parties matters.
One of the worst things universities could do would be to get caught
up in the same kind of bidding wars for facilities and jobs that
states and cities have engaged in.
In the last analysis, the CIC universities are already making vital
contributions to the Midwestern economy and should sustain that
effort. "We have a creative, can-do spirit," Simon said. "We're just
creating for the wrong economy."
See the conference papers.